#5825 Subprocess Error: Backend should be defined in the BACKENDS_MAPPING. Offending backend: tfTraceback (most recent call la

98.128.**1 Posted at: 1 month ago

Subprocess Error: Backend should be defined in the BACKENDS_MAPPING. Offending backend: tf
Traceback (most recent call last):
File "videotrans\task\only_one.py", line 46, in run
File "videotrans\task\_stage_recogn.py", line 62, in recogn
File "videotrans\recognition\__init__.py", line 205, in run
File "videotrans\recognition\_base.py", line 91, in run
File "videotrans\recognition\_whisper.py", line 38, in _exec
File "videotrans\recognition\_whisper.py", line 124, in _faster
File "videotrans\configure\base.py", line 272, in _new_process
videotrans.configure.excepts.VideoTransError: Subprocess Error: Backend should be defined in the BACKENDS_MAPPING. Offending backend: tf

cfg=[TaskCfgVTT]当前工作模式: 翻译视频或音频单视频模式
原始输入文件名: C:/Users/wangy/Downloads/30-Day Drawing Challenge Day 1 Practice Drawing Shapes - Skillshare (720p, h264).mp4,
输出结果保存到文件夹: C:/Users/wangy/Downloads/New folder (3)/30-Day Drawing Challenge Day 1 Practice Drawing Shapes - Skillshare (720p, h264)-mp4,
临时文件夹: C:/Users/wangy/Downloads/win-pyvideotrans-v4.05-0711/tmp/33656/8489a3f9aa
已选清理已存在
未选启用CUDA加速
未选降噪
ASR Channel:faster-whisper(LocalBuilt-in), model_name: large-v3-turbo, 发音语言: Simplified Chinese, 断句方式:默认断句
发音语言和目标语言相同，不翻译字幕
Dubb Channel:Edge-TTS(Free), 角色:Yunyang(Male/CN), 配音语言:Simplified Chinese, 未选二次语音识别
音量:+0%, 语速:+0%, 已选音频加速, 已选视频慢速
字幕: No Subs
已选仅输出mp4
代理地址:

=
system:Windows-10-10.0.26200-SP0
version:v4.05
frozen:True
language:en
root_dir:C:/Users/wangy/Downloads/win-pyvideotrans-v4.05-0711

#1 DeepSeek reply at 1 month ago

This error occurs because the faster-whisper engine (CTranslate2 backend) is trying to use TensorFlow (tf) as its computational backend, but that backend is not registered in the software's built-in mapping. This is usually triggered by a conflicting TensorFlow installation on your system or an incomplete dependency in the bundled environment.

Solutions:

Disable any system-wide TensorFlow
If you have TensorFlow installed globally (e.g., via pip or Anaconda), uninstall it or temporarily rename the TensorFlow package folder. The bundled CTranslate2 expects only PyTorch/ONNX backends, not TensorFlow.
Use openai-whisper (local) instead
In the ASR channel dropdown, select openai-whisper (Local Built-in) with large-v3 model. This avoids CTranslate2 entirely.
Force CPU mode and add environment variable
Create an empty file named CT2_FORCE_CPU=1 in the software root directory, or set the system environment variable CT2_VERBOSE=1 to get more diagnostic info, but the root cause is likely the TensorFlow conflict.
Use a different ASR channel (e.g., Qwen-ASR)
Since your source language is Simplified Chinese, try Qwen-ASR (Local Built-in) which is optimized for Chinese and does not rely on CTranslate2/TensorFlow.

If none of these work, please provide the full log file from logs/ (the 30 lines around the error) for further analysis.

#5825 Subprocess Error: Backend should be defined in the BACKENDS_MAPPING. Offending backend: tfTraceback (most recent call la

1 Replies

Post Your Reply

Related resource